1. Field of the Invention
This invention relates to a distance detecting device and a focus control system utilizing the same.
2. Description of the Prior Art
Various distance detecting devices for automatically detecting the distance to an object and various focus control systems for optical instruments such as cameras or the like for effecting the detection of the in-focus and/or the automatic focus adjustment of the optical system with respect to the object on the basis of the distance signal obtained by said devices have heretofore been proposed, and particularly nowadays, there have been proposed such devices which greatly adopt the recent electronic technique to digitalize the system and thereby digitally carry out the detection of the distance to the object. For example, U.S. Pat. No. 4,004,852 patented Jan. 25, 1977, filed June 30, 1975, title "INTEGRATED AUTOMATIC RANGING DEVICE FOR OPTICAL INSTRUMENTS", granted to Pentecost, proposes a distance detecting device in which, with respect to an object whose distance is to be detected, the images of a first and a second field following different lines of sight and containing said object and different in range are scanned by the use of a line type image sensor while, at the same time, the scanning image element signals obtained at this time are transformed into binary form (quantized) on the basis of a predetermined slice level to obtain continuous M binary image element data regarding the first field image and continuous N (N&gt;M) binary image element data regarding the second field image, and the M image element data regarding the first field image are stored in a first circulation type shift register of M-bit construction while the N image element data regarding the second field image are stored in a second circulation type shift register of N-bit construction, whereafter the detection of bit-by-bit coincidence of the first M bits of stored data of the second shift register with the stored data of the first shift register during one circulation of the stored data of the first and second shift registers and the relative shift of one bit each of the stored data of the second shift register with respect to the stored data of the first shift register are repetitively effected to thereby detect the image portion of the second field image which is most similar to the first field image and know the object distance from the location of this most similar image portion in the second field image.
Also, the assignee of this invention has proposed, in Japanese Patent Application No. 504/1977 (Japanese Laid-open Patent Application Publication No. 85452/1978), a distance detecting device in which when, with respect to an object whose distance is to be detected, the images of a first and a second field following different lines of sight and containing said object and different in range are scanned by a combination of a rockable prism and a pair of light-receiving elements having slits, scanning image element signals obtained by sampling the outputs the light-receiving elements at a predetermined timing are transformed into binary form to obtain M and N continuous binary image element data regarding the first and second field images, respectively, and the M image element data regarding the first field image are stored in a first circulation type shift register of M-bit construction while the first M image element data of the N image element data regarding the second field image are stored in a second circulation type shift register of M-bit construction and the remaining N-M image element data are stored in a third shift register of N-M-bit construction, whereafter the detection of bit-by-bit coincidence between the stored data of the first and second shift registers during one circulation of each of such stored data and the transfer of the stored data of the last bit of the third shift register to the leading bit of the second shift register, namely, the relative shift of one bit each of the stored data of the second and third shift registers with respect to the stored data of the first shift register, are repetitively effected to thereby detect the image portion of the second field image which is most similar to the first field image and know the object distance from the location of this most similar image portion in the second field image.
These proposed distance detecting devices are entirely different from conventional ones and particularly, according to these devices, the scanning image element data obtained by scanning the images are once quantized and then processed, thereby digitally accomplishing the detection of the distance to the object through a digital system and thus, these devices are very excellent in that they can accomplish the detection of the object distance with a high accuracy beyond comparison with conventional devices.
In these distance detecting devices, the relatively complicated processing of quantized image element data as described above is effected for the detection of the object distance and what must be particularly considered in the digital system performing such a complicated operation is making the data processing operation efficient as far as possible and making the system construction efficient. For example, as regards the comparision and processing of the M and N continuous quantized image element data regarding the first and second field images, respectively, according to the device proposed in the aforementioned U.S. Pat. No. 4,004,852, the M image element data regarding the first field image are stored in the first circulation type shift register of M-bit construction while the N image element data regarding the second field image are stored in the second circulation type shift register of N-bit construction, whereafter during circulation of the stored data of these first and second shift registers, until the M stored data of the first shift register make just one circulation, the coincidence or incoincidence between the M stored data of the first shift register and the first M bits of the stored data of the second shift register is detected for each bit and the coincidence number is counted by a counter and, at the point of time whereat the stored data of the first shift register has made just one circulation and returned to their initial state, the driving of the first shift register is stopped and the then count of the counter is compared with the maximum count of the counter hitherto (the count is "0" at the first point of time) (if this count is greater, it is stored as the maximum count), while only the second shift register is further driven by N+M+1 bits to thereby cause the stored data of the second shift register to be shifted by one bit relative to the stored data of the first shift register, whereafter when the stored data of the first and second shift registers are again circulated, during the time until the stored data of the first shift register make just one circulation, the coincidence or incoincidence between the stored data of the first shift register and the first M bits of the stored data of the second shift register is detected for each bit and the coincidence number is again counted by the counter, and such operation is repeated. However, such method of comparison and processing is apparently inefficient in that during the time from after the termination of one circulation and comparison until the next circulation and comparison is effected, the first shift register must remain stopped while only the second shift register must be driven excessively by N-M+1 bits and therefore, before the comparison and processing of all the N data stored in the second shift register is terminated, an excess time represented by at least (M-N+1).times.(M-N).times.(period of clock pulses) is required as compared with the time spent for the actual comparison and processing and accordingly, the time required for the comparison and processing of the data is prolonged and after all, this results in an inconvenience that a long time is required before the object distance is detected. Also, to execute such comparison and processing, it is necessary to repeat such operation in the comparison-processing sequence that, for example, clock pulses are always imparted to the second shift register while M clock pulses are first imparted to the first shift register in synchronism with the start of the circulation of the stored data of the second shift register to thereby cause the stored data of the first shift register to effect one circulation and the ensuing N-M+1 clock pulses are caused to be neglected, whereafter M clock pulses are again imparted to the first shift register to cause the stored data thereof to effect one circulation again, and this leads to a complicated style of drive control for the first shift register in particular, for example, the necessity of providing an exclusive counter for controlling the supply of clock pulses to the first shift register and a logic circuit attendant thereon, which also means complicated construction of the control system for the shift registers.
On the other hand, in the device proposed by the assignee of the present invention (applicant) in Japanese Patent Application No. 504/1977, as far as the comparison and processing of the quantized image element data is concerned, M image element data regarding the first field image are first stored in the first circulation type shift register of M-bit construction while the first M data of the N image element data regarding the second field image are stored in the second circulation type shift register of M-bit construction and the remaining N-M data are stored in the third shift register of N-M-bit construction, whereafter when the stored data of the first and second shift registers are caused to effect one circulation simultaneously, the coincidence or incoincidence between these stored data is detected for each bit and the coincidence number is counted by a counter and, at the point of time whereat the stored data of these first and second shift registers have made just one circulation and returned to their initial state, the driving of the first shift register is temporally interrupted and the then count of the counter is compared with the maximum count of the counter hitherto (the count is "0" at the first point of time) (if this count is greater, it is stored as the maximum count) while, in the meantime, the second and third shift registers are driven by one bit and the data of the leading bit of the third shift register is transferred to the last bit of the second shift register, whereby the stored data of the second and third shift registers are shifted by one bit relative to the stored data of the first shift register, whereafter when the stored data of the first and second shift registers are again caused to effect one circulation simultaneously, the coincidence or incoincidence between the stored data is detected for each bit and the coincidence number is again counted by the counter, and such operation is repeated. According to such method of comparision and processing, during the time from after the termination of one circulation and comparison until the next circulation and comparison is effected, it is only necessary to secure the time required for the comparison of the count of the counter obtained during one circulation and comparison with the maximum count obtained so far, the storage of the count of the counter based on the result of said comparision, and the resetting of the counter thereafter, and moreover, this can be further shortened by parallel-processing the output data of the counter and therefore, as compared with the device of the aforementioned U.S. Pat. No. 4,004,852, there is obtained an advantage that the image element data processing efficiency is much improved and the total time required for the comparison and processing of the image element data is very much shortened. On the other hand, however, in order that such comparison and processing may be executed and that said time may be secured, it is necessary to repeat the operation of temporally cutting off the supply of clock pulses to the first shift register after the terminatior of one circulation of the stored data of the first and second shift registers to interrupt the driving of the first shift register while imparting one excess clock pulse to the second shift register and also imparting one clock pulse to the third shift register to thereby effect relative one-bit shift of the stored data of the second and third shift registers with respect to the stored data of the first shift register and resuming the supply of clock pulses to the first and second shift registers after said necessary time has elapsed after the driving of the first shift register has been stopped, thereby causing the stored data thereof to be circulated and accordingly, there is still left much room for improvement in that the style of drive and control for each shift register is complicated and particularly, the selected supply of clock pulses to each shift register complicates the construction of the control system.
On the other hand, aside from what has been described above, in such a distance detecting device, particularly the digital circuit system thereof is complicated and considerably bulky and therefore, in order to enable such device to be utilized for the focus control system for automatic in-focus detection or automatic focus adjustment in optical instruments such as cameras, it is naturally necessary to make the device into an integrated circuit and thereby make the device compact so that it can easily be incorporated into small instruments, and a particular problem encountered in making the device into an integrated circuit is the number of IC (integrated circuit) elements required. As is well-known, as the number of elements used is smaller, the IC can be made more inexpensive, but on the other hand, if the number of elements used amounts to the order of 10,000, the device must unavoidably be made into an LSI (large scaled integrated circuit) which means a rise in cost. Now, as is generally well-known, even in a digital circuit system of entirely the same function and specification, there is a great difference in number of elements used between the case where it is made into a static type system and the case where it is made into a dynamic type system and by making the digital circuit system into a dynamic type system, the number of elements used can be much reduced as compared with the case where the digital circuit system is made into a static type system. Accordingly, it is much more advantageous to form the digital circuit system as a dynamic type system than to form the digital circuit system as a static type system. On the other hand, however, the dynamic type system encounters its own inconvenience. For example, when the data storing registers are considered, the stored data thereof will be extinguished unless these are always circulated and driven as the data circulation type shift registers by clock pulses, and the clock pulses in this case must have a clock frequency of the order of at least 50 KHz.
Considering the digital circuits in the aforementioned proposed devices from such a point of view, the device of U.S. Pat. No. 4,004,852, as noted above, makes it necessary that during the comparison and processing of the image element data, the first shift register be stopped for a relatively long time after the termination of one circulation and comparison and also, the device proposed by the assignee of the present invention (applicant) in Japanese Patent Application No. 504/1977 makes it necessary that the first and second shift registers be stopped though for a slight time after the termination of one circulation and comparison and accordingly, the construction of the digital circuit systems in these proposed devices are apparently unsuitable to be made into a dynamic type system and thus, in this point as well, there is still left much room for improvement.